Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9035027
Daisuke Kaneishi, J. Leu, Julia O'Donnell, Campbell Affleck, R. Matthew, Andrew McPherson, M. Tomizuka, Hannah S. Stuart
Assistive orthotics have the potential to augment the grasping capabilities of individuals with limited hand functionality. People with a cervical-level spinal cord injury (SCI) lack direct control of semi-flaccid, curled fingers on both hands, which often precludes independent donning and doffing of orthotics, thus limiting their use in daily life. This paper presents a novel orthotic designed to improve hand functionality while facilitating independent daily use for individuals with cervical SCI: the Single-size Semi-soft Assistive Mitten (SSAM). This device utilizes a slim dorsal leaf spring and underactuated cable drive to passively open and actively close the hand, in a way that is robust to variations in hand size. This mitten is intended to improve ease of donning and doffing, as the device is attached to all fingers at once. A prototype is developed based on simulation studies, and grasping performance of the proposed design is validated with five healthy subjects. Successful grasping trials, along with positive user feedback, suggest that the SSAM has a comfortable and intuitive design for long-term, independent use and will enable future studies into feasibility of daily use for people with SCI.
{"title":"Design and Assessment of a Single-size Semi-soft Assistive Mitten for People with Cervical Spinal Cord Injuries","authors":"Daisuke Kaneishi, J. Leu, Julia O'Donnell, Campbell Affleck, R. Matthew, Andrew McPherson, M. Tomizuka, Hannah S. Stuart","doi":"10.1109/Humanoids43949.2019.9035027","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9035027","url":null,"abstract":"Assistive orthotics have the potential to augment the grasping capabilities of individuals with limited hand functionality. People with a cervical-level spinal cord injury (SCI) lack direct control of semi-flaccid, curled fingers on both hands, which often precludes independent donning and doffing of orthotics, thus limiting their use in daily life. This paper presents a novel orthotic designed to improve hand functionality while facilitating independent daily use for individuals with cervical SCI: the Single-size Semi-soft Assistive Mitten (SSAM). This device utilizes a slim dorsal leaf spring and underactuated cable drive to passively open and actively close the hand, in a way that is robust to variations in hand size. This mitten is intended to improve ease of donning and doffing, as the device is attached to all fingers at once. A prototype is developed based on simulation studies, and grasping performance of the proposed design is validated with five healthy subjects. Successful grasping trials, along with positive user feedback, suggest that the SSAM has a comfortable and intuitive design for long-term, independent use and will enable future studies into feasibility of daily use for people with SCI.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132045988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/humanoids43949.2019.9035041
{"title":"RAL Papers Presented at Humanoids2019","authors":"","doi":"10.1109/humanoids43949.2019.9035041","DOIUrl":"https://doi.org/10.1109/humanoids43949.2019.9035041","url":null,"abstract":"","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"963 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132445266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9034997
Sumin Park, Jaeheung Park
A robot foot can slip when the horizontal shear force acting on the foot exceeds the frictional force between the foot and the ground. In the linear inverted-pendulum (LIP) model, the vertical height of the center of mass (COM) is kept constant, and the vertical force is always equal to the gravitational force at any walking speed. However, the horizontal force increases upon increasing the walking speed. This restriction on the vertical force in the LIP model can cause the robot foot to slip at fast walking speeds, as the horizontal force can exceed the frictional force, which is proportional to the vertical force. In this study, we present an optimization method to generate vertical COM motion to maintain the utilized coefficient of friction (uCOF) less than the available coefficient of friction between the foot and the ground, and to minimize the mechanical work of the COM. Vertical motions at various speeds are generated using the proposed optimization method. Subsequently, the generated COM motion patterns are used as reference trajectories of the COM in robot simulation. Optimization and simulation results demonstrate that the mechanical work and uCOF decrease because of the vertical motion. This study suggests a way to generate slip-safe and energy-efficient COM patterns, which, in turn, overcome the limitations of the LIP model by adding vertical COM motion.
{"title":"Vertical COM Motion Generation to Reduce Slipping and Mechanical Work during Walking","authors":"Sumin Park, Jaeheung Park","doi":"10.1109/Humanoids43949.2019.9034997","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9034997","url":null,"abstract":"A robot foot can slip when the horizontal shear force acting on the foot exceeds the frictional force between the foot and the ground. In the linear inverted-pendulum (LIP) model, the vertical height of the center of mass (COM) is kept constant, and the vertical force is always equal to the gravitational force at any walking speed. However, the horizontal force increases upon increasing the walking speed. This restriction on the vertical force in the LIP model can cause the robot foot to slip at fast walking speeds, as the horizontal force can exceed the frictional force, which is proportional to the vertical force. In this study, we present an optimization method to generate vertical COM motion to maintain the utilized coefficient of friction (uCOF) less than the available coefficient of friction between the foot and the ground, and to minimize the mechanical work of the COM. Vertical motions at various speeds are generated using the proposed optimization method. Subsequently, the generated COM motion patterns are used as reference trajectories of the COM in robot simulation. Optimization and simulation results demonstrate that the mechanical work and uCOF decrease because of the vertical motion. This study suggests a way to generate slip-safe and energy-efficient COM patterns, which, in turn, overcome the limitations of the LIP model by adding vertical COM motion.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133344091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9034994
Kento Kawaharazuka, Kei Tsuzuki, Moritaka Onitsuka, Yuya Koga, Yusuke Omura, Yuki Asano, K. Okada, Koji Kawasaki, M. Inaba
The musculoskeletal humanoid can move well under environmental contact thanks to its body softness. However, there are few studies that actively make use of the environment to rest its flexible musculoskeletal body. Also, its complex musculoskeletal structure is difficult to modelize and high internal muscle tension sometimes occurs. To solve these problems, we develop a muscle relaxation control which can minimize the muscle tension by actively using the environment and inhibit useless internal muscle tension. We apply this control to some basic movements, the motion of resting the arms on the desk, and handle operation, and verify its effectiveness.
{"title":"Reflex-Based Motion Strategy of Musculoskeletal Humanoids under Environmental Contact Using Muscle Relaxation Control","authors":"Kento Kawaharazuka, Kei Tsuzuki, Moritaka Onitsuka, Yuya Koga, Yusuke Omura, Yuki Asano, K. Okada, Koji Kawasaki, M. Inaba","doi":"10.1109/Humanoids43949.2019.9034994","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9034994","url":null,"abstract":"The musculoskeletal humanoid can move well under environmental contact thanks to its body softness. However, there are few studies that actively make use of the environment to rest its flexible musculoskeletal body. Also, its complex musculoskeletal structure is difficult to modelize and high internal muscle tension sometimes occurs. To solve these problems, we develop a muscle relaxation control which can minimize the muscle tension by actively using the environment and inhibit useless internal muscle tension. We apply this control to some basic movements, the motion of resting the arms on the desk, and handle operation, and verify its effectiveness.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116161425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9035032
M. Nicolescu, Natalie Arnold, Janelle Blankenburg, David Feil-Seifer, S. Banisetty, M. Nicolescu, Andrew H. Palmer, Thor Monteverde
This paper presents a novel approach to robot task learning from language-based instructions, which focuses on increasing the complexity of task representations that can be taught through verbal instruction. The major proposed contribution is the development of a framework for directly mapping a complex verbal instruction to an executable task representation, from a single training experience. The method can handle the following types of complexities: 1) instructions that use conjunctions to convey complex execution constraints (such as alternative paths of execution, sequential or non-ordering constraints, as well as hierarchical representations) and 2) instructions that use prepositions and multiple adjectives to specify action/object parameters relevant for the task. Specific algorithms have been developed for handling conjunctions, adjectives and prepositions as well as for translating the parsed instructions into parameterized executable task representations. The paper describes validation experiments with a PR2 humanoid robot learning new tasks from verbal instruction, as well as an additional range of utterances that can be parsed into executable controllers by the proposed system.
{"title":"Learning of Complex-Structured Tasks from Verbal Instruction","authors":"M. Nicolescu, Natalie Arnold, Janelle Blankenburg, David Feil-Seifer, S. Banisetty, M. Nicolescu, Andrew H. Palmer, Thor Monteverde","doi":"10.1109/Humanoids43949.2019.9035032","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9035032","url":null,"abstract":"This paper presents a novel approach to robot task learning from language-based instructions, which focuses on increasing the complexity of task representations that can be taught through verbal instruction. The major proposed contribution is the development of a framework for directly mapping a complex verbal instruction to an executable task representation, from a single training experience. The method can handle the following types of complexities: 1) instructions that use conjunctions to convey complex execution constraints (such as alternative paths of execution, sequential or non-ordering constraints, as well as hierarchical representations) and 2) instructions that use prepositions and multiple adjectives to specify action/object parameters relevant for the task. Specific algorithms have been developed for handling conjunctions, adjectives and prepositions as well as for translating the parsed instructions into parameterized executable task representations. The paper describes validation experiments with a PR2 humanoid robot learning new tasks from verbal instruction, as well as an additional range of utterances that can be parsed into executable controllers by the proposed system.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121553253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9035028
M. Harant, M. Millard, N. Šarabon, K. Mombaur
Spinal exoskeletons can reduce the risk of low-back pain by decreasing the back muscle activity and the spinal compression forces of users during heavy lifting tasks. Model-based simulation and optimization are helpful tools to support the design of exoskeletons reducing the number of prototypes iterations and testings. In this paper, we present a comparison of different cost functions based on modeling and optimization techniques to determine proper actuator characteristics for an active spinal exoskeleton supporting stoop-lifts of a 10 kg box. Using motion recordings of five different subjects and additional anthropometric measurements, we created subject-specific musculoskeletal models. A corresponding parameterized model of an exoskeleton with passive and active components was created and combined with the human models. The spring characteristics and the torque profiles of the exoskeleton are optimized for various objectives which consist of a term for tracking the recordings and an additional term from a set of cost functions to reduce the load on the subjects. User comfort is guaranteed by appropriate interaction force limits. The results show that all cost functions reduced significantly the human torque loads. However, they result in different amounts and distributions of the load reduction as well as different contributions from the passive and active components of the exoskeleton. They also yield different actuation patterns of the human model performing the stoop-lift. The analysis of the effects of cost functions in this study highlights the importance of selecting an appropriate cost function for optimization.
{"title":"Cost function evaluation for optimizing design and actuation of an active exoskeleton to ergonomically assist lifting motions","authors":"M. Harant, M. Millard, N. Šarabon, K. Mombaur","doi":"10.1109/Humanoids43949.2019.9035028","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9035028","url":null,"abstract":"Spinal exoskeletons can reduce the risk of low-back pain by decreasing the back muscle activity and the spinal compression forces of users during heavy lifting tasks. Model-based simulation and optimization are helpful tools to support the design of exoskeletons reducing the number of prototypes iterations and testings. In this paper, we present a comparison of different cost functions based on modeling and optimization techniques to determine proper actuator characteristics for an active spinal exoskeleton supporting stoop-lifts of a 10 kg box. Using motion recordings of five different subjects and additional anthropometric measurements, we created subject-specific musculoskeletal models. A corresponding parameterized model of an exoskeleton with passive and active components was created and combined with the human models. The spring characteristics and the torque profiles of the exoskeleton are optimized for various objectives which consist of a term for tracking the recordings and an additional term from a set of cost functions to reduce the load on the subjects. User comfort is guaranteed by appropriate interaction force limits. The results show that all cost functions reduced significantly the human torque loads. However, they result in different amounts and distributions of the load reduction as well as different contributions from the passive and active components of the exoskeleton. They also yield different actuation patterns of the human model performing the stoop-lift. The analysis of the effects of cost functions in this study highlights the importance of selecting an appropriate cost function for optimization.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125251747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9035079
Kotaro Funakoshi, Ryota Yamagami, S. Sugano, Mikio Nakano
Response obligation is whether a spoken dialogue system should react to an input sound. This paper focuses on the false negative errors in response obligation estimation (ROE) that are displayed as the system's neglect of its users. When the users repeat after the system ignores their speech, ROE will likely fail again because the repeated input is similar to the previous input. Therefore, we propose an improved ROE method that considers users' repetitions. First, we show that a simple concatenation of ROE and repetition features is better than two other integration architectures. Then, we propose a modified random forest algorithm that incorporates human domain knowledge. The effectiveness is demonstrated with simulated repetitions as a 7.6-point gain from the baseline.
{"title":"Response Obligation Estimation That Considers Users' Repetitive Utterances using Knowledge-Guided Random Forest","authors":"Kotaro Funakoshi, Ryota Yamagami, S. Sugano, Mikio Nakano","doi":"10.1109/Humanoids43949.2019.9035079","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9035079","url":null,"abstract":"Response obligation is whether a spoken dialogue system should react to an input sound. This paper focuses on the false negative errors in response obligation estimation (ROE) that are displayed as the system's neglect of its users. When the users repeat after the system ignores their speech, ROE will likely fail again because the repeated input is similar to the previous input. Therefore, we propose an improved ROE method that considers users' repetitions. First, we show that a simple concatenation of ROE and repetition features is better than two other integration architectures. Then, we propose a modified random forest algorithm that incorporates human domain knowledge. The effectiveness is demonstrated with simulated repetitions as a 7.6-point gain from the baseline.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130865527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9034998
Lipeng Chen, Luis F. C. Figueredo, M. Dogar
This paper addresses the problem of sequential manipulation planning to keep an object stable under changing external forces. Particularly, we focus on using object-environment contacts. We present a planning algorithm which can generate robot configurations and motions to intelligently use object-environment, as well as object-robot, contacts, to keep an object stable under forceful operations such as drilling and cutting. Given a sequence of external forces, the planner minimizes the number of different configurations used to keep the object stable. An important computational bottleneck in this algorithm is due to the static stability analysis of a large number of configurations. We propose a containment relationship between configurations, to prune the stability checking process.
{"title":"Manipulation Planning Using Environmental Contacts to Keep Objects Stable under External Forces","authors":"Lipeng Chen, Luis F. C. Figueredo, M. Dogar","doi":"10.1109/Humanoids43949.2019.9034998","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9034998","url":null,"abstract":"This paper addresses the problem of sequential manipulation planning to keep an object stable under changing external forces. Particularly, we focus on using object-environment contacts. We present a planning algorithm which can generate robot configurations and motions to intelligently use object-environment, as well as object-robot, contacts, to keep an object stable under forceful operations such as drilling and cutting. Given a sequence of external forces, the planner minimizes the number of different configurations used to keep the object stable. An important computational bottleneck in this algorithm is due to the static stability analysis of a large number of configurations. We propose a containment relationship between configurations, to prune the stability checking process.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125180579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9035001
Diogo Almeida, E. Cansizoglu, Radu Corcodel
We present an interactive perception system that enables an autonomous agent to deliberately interact with its environment and produce 3D object models. Our system verifies object hypotheses through interaction and simultaneously maintains 3D SLAM maps for each rigidly moving object hypothesis in the scene. We rely on depth-based segmentation and a multigroup registration scheme to classify features into various object maps. Our main contribution lies in the employment of a novel segment classification scheme that allows the system to handle incorrect object hypotheses, common in cluttered environments due to touching objects or occlusion. We start with a single map and initiate further object maps based on the outcome of depth segment classification. For each existing map, we select a segment to interact with and execute a manipulation primitive with the goal of disturbing it. If the resulting set of depth segments has at least one segment that did not follow the dominant motion pattern of its respective map, we split the map, thus yielding updated object hypotheses. We show qualitative results with a Fetch manipulator and objects of various shapes, which showcase the viability of the method for identifying and modelling multiple objects through repeated interactions.
{"title":"Detection, Tracking and 3D Modeling of Objects with Sparse RGB-D SLAM and Interactive Perception","authors":"Diogo Almeida, E. Cansizoglu, Radu Corcodel","doi":"10.1109/Humanoids43949.2019.9035001","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9035001","url":null,"abstract":"We present an interactive perception system that enables an autonomous agent to deliberately interact with its environment and produce 3D object models. Our system verifies object hypotheses through interaction and simultaneously maintains 3D SLAM maps for each rigidly moving object hypothesis in the scene. We rely on depth-based segmentation and a multigroup registration scheme to classify features into various object maps. Our main contribution lies in the employment of a novel segment classification scheme that allows the system to handle incorrect object hypotheses, common in cluttered environments due to touching objects or occlusion. We start with a single map and initiate further object maps based on the outcome of depth segment classification. For each existing map, we select a segment to interact with and execute a manipulation primitive with the goal of disturbing it. If the resulting set of depth segments has at least one segment that did not follow the dominant motion pattern of its respective map, we split the map, thus yielding updated object hypotheses. We show qualitative results with a Fetch manipulator and objects of various shapes, which showcase the viability of the method for identifying and modelling multiple objects through repeated interactions.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128820595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/Humanoids43949.2019.9035063
Michael J. Mathew, Saif Sidhik, M. Sridharan, M. Azad, Akinobu Hayashi, J. Wyatt
During the initial trials of a manipulation task, humans tend to keep their arms stiff in order to reduce the effects of any unforeseen disturbances. After a few repetitions, humans perform the task accurately with much lower stiffness. Research in human motor control indicates that this behavior is supported by learning and continuously revising internal models of the manipulation task. These internal models predict future states of the task, anticipate necessary control actions, and adapt impedance quickly to match task requirements. Drawing inspiration from these findings, we propose a framework for online learning of a time-independent forward model of a manipulation task from a small number of examples. The measured inaccuracies in the predictions of this model dynamically update the forward model and modify the impedance parameters of a feedback controller during task execution. Furthermore, our framework includes a hybrid force-motion controller that provides compliance in particular directions while adapting the impedance in other directions. These capabilities are evaluated on continuous contact tasks such as pulling non-linear springs, polishing a board, and stirring porridge.
{"title":"Online Learning of Feed-Forward Models for Task-Space Variable Impedance Control","authors":"Michael J. Mathew, Saif Sidhik, M. Sridharan, M. Azad, Akinobu Hayashi, J. Wyatt","doi":"10.1109/Humanoids43949.2019.9035063","DOIUrl":"https://doi.org/10.1109/Humanoids43949.2019.9035063","url":null,"abstract":"During the initial trials of a manipulation task, humans tend to keep their arms stiff in order to reduce the effects of any unforeseen disturbances. After a few repetitions, humans perform the task accurately with much lower stiffness. Research in human motor control indicates that this behavior is supported by learning and continuously revising internal models of the manipulation task. These internal models predict future states of the task, anticipate necessary control actions, and adapt impedance quickly to match task requirements. Drawing inspiration from these findings, we propose a framework for online learning of a time-independent forward model of a manipulation task from a small number of examples. The measured inaccuracies in the predictions of this model dynamically update the forward model and modify the impedance parameters of a feedback controller during task execution. Furthermore, our framework includes a hybrid force-motion controller that provides compliance in particular directions while adapting the impedance in other directions. These capabilities are evaluated on continuous contact tasks such as pulling non-linear springs, polishing a board, and stirring porridge.","PeriodicalId":404758,"journal":{"name":"2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124558048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：